From 0d6526f82c3cdefcca47f73f5fc08dc6f335eac6 Mon Sep 17 00:00:00 2001 From: Tim Chen Date: Mon, 21 Sep 2026 17:37:22 -0700 Subject: sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling bug alb_break_llc() decides whether to break LLC preference during active load balance. It does so by testing that every runnable fair task on the source rq prefers its LLC: env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable But the two counters cover different sets. nr_pref_llc_running is updated in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued, so it follows queued tasks. h_nr_runnable is updated in set_delayed()/ clear_delayed() and drops delay-dequeued tasks. So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks, alb_break_llc() returns false, and active balance is free to pull a task off its preferred LLC. Active balance only moves runnable tasks, and this is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB skips the per-task test in can_migrate_task(). The runnable set is the one we want. Fix it on the counter side. A task should be counted in nr_pref_llc_running exactly while it is both queued on its preferred LLC (pref_llc_queued) and runnable (!sched_delayed). Define that membership once in task_pref_llc_runnable(), and adjust the counter only through pref_llc_running_inc()/pref_llc_running_dec() from the four sites that change either input: account_llc_enqueue(), account_llc_dequeue(), set_delayed() and clear_delayed(). Gating every update on the same predicate keeps the delay, wake and dequeue paths from double-counting or underflowing; see the comments at those sites for the ordering. nr_llc_running and sd->llc_counts are not touched and stay on queued semantics. Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC") Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/ Reported-by: Zhan Xusheng Suggested-by: Chen Yu Signed-off-by: Tim Chen Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Kayra Cizmeci Cc: # v7.2.x Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com --- arch/powerpc/sysdev/mpic.h | 60 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+) create mode 100644 arch/powerpc/sysdev/mpic.h (limited to 'arch/powerpc/sysdev/mpic.h') diff --git a/arch/powerpc/sysdev/mpic.h b/arch/powerpc/sysdev/mpic.h new file mode 100644 index 000000000..bb460ff57 --- /dev/null +++ b/arch/powerpc/sysdev/mpic.h @@ -0,0 +1,60 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +#ifndef _POWERPC_SYSDEV_MPIC_H +#define _POWERPC_SYSDEV_MPIC_H + +/* + * Copyright 2006-2007, Michael Ellerman, IBM Corporation. + */ + +#ifdef CONFIG_PCI_MSI +extern void mpic_msi_reserve_hwirq(struct mpic *mpic, irq_hw_number_t hwirq); +int __init mpic_msi_init_allocator(struct mpic *mpic); +int __init mpic_u3msi_init(struct mpic *mpic); +#else +static inline void mpic_msi_reserve_hwirq(struct mpic *mpic, + irq_hw_number_t hwirq) +{ + return; +} + +static inline int mpic_u3msi_init(struct mpic *mpic) +{ + return -1; +} +#endif + +#if defined(CONFIG_PCI_MSI) && defined(CONFIG_PPC_PASEMI) +int __init mpic_pasemi_msi_init(struct mpic *mpic); +#else +static inline int mpic_pasemi_msi_init(struct mpic *mpic) { return -1; } +#endif + +extern int mpic_set_irq_type(struct irq_data *d, unsigned int flow_type); +extern void mpic_set_vector(unsigned int virq, unsigned int vector); +extern int mpic_set_affinity(struct irq_data *d, + const struct cpumask *cpumask, bool force); +extern void mpic_reset_core(int cpu); + +#ifdef CONFIG_FSL_SOC +extern int mpic_map_error_int(struct mpic *mpic, unsigned int virq, irq_hw_number_t hw); +void __init mpic_err_int_init(struct mpic *mpic, irq_hw_number_t irqnum); +int __init mpic_setup_error_int(struct mpic *mpic, int intvec); +#else +static inline int mpic_map_error_int(struct mpic *mpic, unsigned int virq, irq_hw_number_t hw) +{ + return 0; +} + + +static inline void mpic_err_int_init(struct mpic *mpic, irq_hw_number_t irqnum) +{ + return; +} + +static inline int mpic_setup_error_int(struct mpic *mpic, int intvec) +{ + return -1; +} +#endif + +#endif /* _POWERPC_SYSDEV_MPIC_H */ -- cgit v1.3.1